dermot.shane

almost three years later!

thanks Matthew

i have not seen another FCX project since that one,  in the three years since the show from hell  i've graded 30+ features, all but two from MC, and the outliers were Pp

i've cut-n-pasted your notes to a word doc, along with links to EDL-X  and  X2CC  - could be super helpful to have them at hand

typcialy,  i do the conform,  and typicaly it's problem free,  if there's any significant issues i'll send the project on to an assist to take on the compound clip -> media pool -> conform clip shuffle

in the last three years i've mothballed the DS  stations,  parked the Flame, hae not had a reason to turn it on in over a year, and am mainly working in Resolve, with two shows done in Baselight

thanks Rakesh, i've had great support from HP in the past, but that was all about hardware issues, our local support in Vancouver Canada is amazing top to tail, but not able to help on this file

now an update:
i got RGS working, painful and tedious, but working...

after setting RGS to easy login, restarting the machine (not mentioned in the manual) i got it to work inside the firewall at the studio, the machines need a static IP, already had that, but worth mentioning

then we needed to set up a VPN to get through our firewall from my suite at home, and my MBP

good stuff after an hour with it:

-it's fast  - as in i see 24Fps playback without dropped frames playing Resolve's UI on my laptop in a local cafe with a crap connection, far, far better than Teamviewer,  have both on the MBP so could run them one after an other, the diffrence is quite noticable

-far better image quality than Teamviewer can offer as well, by a long shot

bad stuff:

- supremely user unfriendly to set up

- need to manualy enter IP addy, machine name, and password EVERY time, every single time...it has no list of approved / safe machines

- no scaling options - it’s only 1:1 making navigation a needless pita with a smaller screen - yes Mr MBP, i’m looking at you…

- need to give anyone  you want to share a screen with your VPN addy, machine name and password! Ekkkkk! like that’s gonna happen…

 in the last  three days on my machine i’ve had clients approveing VFX / editor sorting conform issues / Assist transferring DRP and checking  - all logged on with TeamViewer and every time i reset the password after they logged out

i’m not going to re-set the machine name and password endlessly, not to mention i have to restart the machine for windoze to accept a user name change every time... ouch....

HP support got back to me, they are esclateing, but that was before i got it going, still going to ask question to make sure i'm understanding the login process, and see if there's a way to have a cleint/assist/ VFX artist whomever log on without needing to give them access to the machine permently

Jason Myres @ LGG suggested looking at "NoMachine" https://www.nomachine.com/

so i'm leaning toward haveing both NoMachine for cleints to log on, probbaly on the z800, and me to use from my laptop so it can scale, and RGS between machines both at the studio and at home, where i'm comfortable that i can keep the IP / user name / password info under control

any machine can be accessed from the z800 via RGS so the z800 serves as a hub

and a one year subscription to NoMachine costs less than a month of TeamViewer

one thing's for sure, i'll be dropping TeamViewer when the yearly contract runs out....  the same resources could pay for a year of Kognat, NoMachine, upgrade BCC to current and more....

has anyone got it working?

i wanted to test it, installed sender/receiver on four HP z class machines, and receiver on my MBP, set all senders to easy logon, and get the same error message on all machines


as i've entered correct ip / machine name / password, and made sure RGS passes through the firewall, i'm thinking it might be #2...
i've read the manual, and no help that i can see there
i've reached out to HP support forum, no answers
i've emailed support and they sent me the manual, and have ghosted me since
i've asked local HP tech's and they deal with hardware only

TeamViewer yearly contract is up at the end of the month, i'd love to eval RGS, but can't get connected to even take a peek at it

any help appreacated

+1 on the Drastic 4k scopes

there's a few choices, this is the one i went with:

https://www.drastic.tv/productsmenu-56/test-and-measurement/4kscope

i used BMD's Ultrascope before the Drastic scope, they run on the same minimilist hardware / both have error logging

i do love the vector scope display in ScopeBox, but i love the confidence of sending clean masters even more....

i think there's a gulf between what makes for a pass in a TVC, and what makes for a pass from a "class A faclity".. ie: Technicolor, Deluxe, Photokem, and a pass from a "class A faclity" is what is in the delivery spec's i deal with typicaly

a delviery service checkbox does not apply to QC reports unfortuantly, either you have program content or you have a QC fail, full stop, no checkbox to check here

the most common QC fail is chroma excursions, even tho it's based on an out dated standard, it's written into the delivery spec's and the QC facility have to make sure the show meets that, no wiggle room, and no admiting current reality exists....

end game is to make sure to meet the delivery spec's and if "broadcast safe" is the spec, then you have to have content in every frame, and you have to control chroma excursions, and not live in the hope that the delivery partner has a checkbox to check, as not all do....

4 minutes ago, dermot.shane said:

 

3 minutes ago, dermot.shane said:

 

4 hours ago, Nico Wieseneder said:

Thx a lot for your help, Dermot! Do you know why the „Broadcast Safe Button“ does compress/desat too much even if it would be also broadcast safe with less compression/desat?

i certainly will be passable with a more subtle approach

in Resolve i use the gamut OFX, and if it's touching other parts of the image without issues, put it inside a qualifier sometimes

less common is to use log highlights/shadows + softclip per channel + a node set to HSV with ch1&3 bypassed to control sat in shadows/highlights with the log wheels... those were the tools of choice a few years ago, when that was all that was avb, i find gamut mapping is faster and creates less collatrial damage

i  would love to have the good stuff in ScopeBox, but the drawback of no live error logging rules it out for my workflow

i have a  somewhat diffrent experience from Amada's

i deliver to distb's who run shows through QC process, and yes they care very much about chroma excursions etc.... local tv deliveries have variable standards to meet, but i treat them the same as going to  a top tier streaming service

if you have a chroma excursion in your master timeline, then you will have that in a delvierable of any sort, ProRez, DPX whatever...

if the broadcaster does not care, then happiness abounds

if the broadcaster runs the show through a error logging software, then you will have to care, and repair to meet deliverable spec, or the broadcaster will fix it and send the bill to the producer, but it will not run if it fails QC

should note i have a fair amount of buisness fixing other colorist's QC fails for a network, when the filmmaker and the colorist cannot deliver a master that passes QC

also note running QC costs the producer around 5k USD for a feature, you will not want to be responcable for a second fail.. the first one is allowed for, a second fail is usualy not budgeted for, not in terms of delvierables deadlines, or charges...

a second fail is a good way to lose a client, it costs the producer too much, and end game is it's cheaper to pay more up front to get a QC pass and deliver on time... bank charges on the money tied up in a film  near delivery are substaintial, the producer does not get their MG paid out until the QC report says "PASS"... like a new BMW's worth of bank charges every month....

so yea, i watch my error logging like a hawk looking for dinner.... and i pass QC on the first go usualy, the only fails i've had recently have been for doggy VFX, and that's out of my hands

i flag any area's of technical concern, and what i have done to  minimise them, and area's where a creative choice has been made that effects image quality,and send these notes to both the producer's and the QC facility, that helps getting the show through cleanly as well

this is an area where experience and a proven track record can really help when pitching your services to a producer, and possibaly allow you to account for a higher day rate when billing,and it's a hard train to get onto as typicaly you need to have the street cred to even get the gig, or charge so little that there's room in the post budget to fix the errors, and pay the bank charges when the show is not delviered .. a three day delay to get through a second round of QC and the calander flipping to a new month during that time means the producer just threw away the price of a new car for your errors... you really do not want to be that guy...

and there's some things that are not really obvious that will create fails.. the first frame of a  fade up from back cannot be 0% black, QC will fault the show for the content being too short... i typicaly put a bit of noise there so something shows up on a scope.... something i would not expect local TVC's to have an issue with for example

best bet is not to do that in the first place

instead grade in 709,

and hire a gradeing theatre for a trim pass

you likely will not gain anything with P3's wider gamut anyway

i use scopes with real time error logging, and keep a close eye on those

never , ever, never use the "broadcast safe" lut on export, never....

if you work in ACES chk out Paul Dore's PFE.. an elegant and flexiable answer

if you can run 1/4rez then the disks subsystem is fast enough for any debayer setting

debayer will soon be on GPU, i would wait a week or more, and see if the expected gains are real, if so.. a GPU upgrade may be the better choice

it's  on all windoze machines, no matter the cpu/gpu etc etc

super e-z workaround is to make the folder with the os, does not effect me much as i use a fixed folder structure already, so any folder i need is more than likely already there

dev's know about it, hopefully it gets sorted sooner or later, but there's bigger bugs to sort first IMHO, like ACES on all os's after 15.2.2 needs some lurve

they know about that as well....

ACEScct....

not really what we had to deal with, but thanX

the XML's we tried would show the same issues even when opened in another FCX

as noted above Nucoda, FC7 and Pp were also tried, with the same resualt

we did try X-2-7 to get the XML into 7, no joy

i have since conformed three more features in Resolve with out any issues, two from MC, one from Pp

this is more like the kinda typical workflow in 2019

my machines are not an issue either, unlike the trashcan in the BMD thread, i run HP workstations with dual Xeons, Dual GPU's, 128gig sys ram, 64Tb NAS on 10gigE + 32Tb  each SAS local storage....  solid and supported machines

thnx Brian,

that link is very VERY basic, and was all followed to the letter in the first place, as it could be lifted from my decades old cheat sheet that i send to any editor

basic good clean workflows are needed, but really no one involved in this is a beginner, the network's online editor was a scanner op in 2006, we worked on projects together back then i was gradeing his scans, one example of the depth the crew involved have under their collective belt

i've been conforming shows since the days of 2" machines with an Ampex ACE, and convergent s90 used for offline, read the late 1970's....

it takes a fair bit to screw up something that i can't un-screw up by now, we even got this sorted, mainly by eyematching with the editors help as he could read out the correct tc/file name from FCX, shot-by-shot... all 90 min

it really comes back to FCX, and the lack of support -or- information on what the weak links are, how to sort or avoid

far beyond the basic's

but that's a solid document, well worth haveing peak if you are not used to doing a handover to sound and picture, every union assist are super solid on this file, and that's who i normaly get handover from

we tired that, but unfortunatly garbage in ( the XML) = garbage out, it did the same job as exporting an EDL from Resolve or Nucoda, all avenues we could think of were tried,  and personaly i was not hopeful on that one anyway, so not disapointed 😉

and many thanks to you, Brad, Thomas, and everyone else who tried to help

moveing forward, if there is ever a bullet proof FCX editor's cheat sheet floating around ti would be loverly to have access to that, so "issues" could be headed off before they become "issues"...

something beyond the endless "press this button to make an XML, and everything will be perfect" drivel spread everywhere on the interwebs, it seems to be a disservice to spread the message that it works when in reality it is far from working beyond a very simple demo timeline

i really was hopeing for some thoughts from the editor, but he's done with FCX, not interested in touching it again, so has zero intrest in sorting the weak links for future use, and has his hands full catching up on all the stuff he dropped to sort this in the last two weeks, he did not budget his time for this either

the issues center on not being able to de-compose the mutlicam, and de-composing compund clips did not actualy de-compose all the clips, only some, and when it did de-compose, it did it incorrectly... all this is about FCX and not about creating the XML or any system that injests the XML not intrepreting it correctly

Apple support was somewhere just slightly less than zero, he got a canned reply and that's all...

on the good side this film is going to get some visability, opening night gala film at a rather well known festival in the fall, so i'm kinda reluctant to mention the name and have it turn up in a google search.. and the editor did an awesome job crafting the story

the only failing was not allowing time and budget to get a clean turnover, the sched was pushed back from gradeing just after Christmas to piclok on Feb10, but the hard deadline was still Feb 20 to hand over my CTM to finishing....

would be possiable without drama with Pp, MC on and on...  was barely possiable, and at that only with massive drama for both sound and picture

On 2/6/2019 at 12:14 PM, dermot.shane said:

appearing in Resolve with -99.00.00.00 timecode and media offline

appearing in Resolve with 00.00.00.00 timecode and media ofline

appearing in Resolve with wrong media

appearing in Resolve with black

appearing in Resolve with the correct media, but out of sync

correct but showing what i know are r3d files as 1920x1080 with no access to the raw tab

could be any one of the above seemingly randomly

so yea, compound/mutilcam clips come into Resolve, Nucoda, Pp, FC7...

they come in with all the issues listed above

mutiple software tried

mutiple finishing artists / online editors tried

no one could get anything but garbage out of the XML

the editor could not get it any better

even importing the XML into a diffrent FCX install failed to link to the cam orig.... like really.... it can't even open it's own XML.. smart folks tried

i asked the editor to write me a cheat sheet for future FCX editors, but they only this he would write is "don't use FCX"

again , really, literaly the best cheat sheet he could come up with is not to use it, and he's the one driveing it

no idea what version of FCX he had

we tried current Pp, Nucoda 2018.2 and Resolve 15.2.2

the network's post sup and crew tried Pp, Resolve 15.2.3, FC7 and a seperate current FCX they keep around for times like this

the editor tried current Pp and FC7

all failed

a perfect 100% fail rate

this one is now gone to QC

i'm thrilled if a great show appears cut on FCX, it's a great show then be happy the editor had the tools they wanted and did great work, i'm pretty offline edit agnostic

would be great to have some real world probelem sorting, not yet another boring, tired, basic  "geee wizz kidz, this is how you export an xml...." bumph

and there is loads of resources should Avid or Pp project go off the rails, the resources for sorting FCX trainwrecks is thin on the ground... and that seems to be  a weakness of the FCX community,

the editor asked anyone and everyone he could, i also asked the smart folks here... i really appreacate the help offered here, and i passed it on to the editor who tried every suggestion

but wowzers, one more basic, boring, out of touch with reality tutorial on clicking a button is prolly not how i want to use the time i have

the ripple training looks as the very basic functions assumeing a clean and simple timeline in FCX.. not anything near the real  world of complex timeline with compound and mutlicam clips

this show is done now, the outfall of this is:

- the producer could have bought a new car cheaper than paying pro's to fix this,

- the garbage XML also affected the mix in a major way.. they were just a hosed as i was

- the producer's were paying for a mix theatre and gradeing suite to do assist work, hours were huge as we were under a tight deadline so no chance to do the sensible thing, and go back to Pp or MC and ask an assist to eye match the whole 90 min, then issue an XML/AAF/OMF that was not garbage

- the network who commisioned it have now banned FCX from any project submitted by filmakers, they will have to submit a XML from Pp. FC7,Resolve or AAF from MC  from now on, how they get there is up to them, but the qt and XML cannot be issued from FCX timeline anymore

- the editor will never cut anything on FCX again, as he lost huge hours, and other gigs while he dropped everything to sort this

- and i am sticking to my prefrence...  send and FCX XML to an assist and pay them to sort the mess before it enters the world of color and mix, and be prepared to pay for their time, basicly add $3k and a week to the  end of any FCX project

i don't care what tools an editor uses, they are the one's telling the story,so as long as the producer's are prepared to pay for FCX and allow time in their sched... i see no reason why it should not be used if the editor can tell their story best with it

just hope that production is aware of the expense at the back end

i'm not sure i understand the issue

did you create custom mapping and then recall the default, and now cannot find the custom mapping to re-load?

one thought, just cause i use L*a*b alot.. on the "L*a*b sat" PG, set pivot to .5, that gives more range to the effect using contrast to push it to 11 should it be needed;

Quote

so what is the problem?

appearing in Resolve with -99.00.00.00 timecode and media offline

appearing in Resolve with 00.00.00.00 timecode and media ofline

appearing in Resolve with wrong media

appearing in Resolve with black

appearing in Resolve with the correct media, but out of sync

correct but showing what i know are r3d files as 1920x1080 with no access to the raw tab

could be any one of the above seemingly randomly

and thanks for taking the time to help the editor out

help needed!

i have a feature cut in FCX conformed last night

there's a zillion "mutil-cam" clips in there

the editor has de-composed the compound clips but cannot figure out how to de-compose these mutlicam clips

any advice i can pass on?